> ## Documentation Index
> Fetch the complete documentation index at: https://docs.galtea.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Runs Service

> Runs Service API methods in the Galtea SDK

The Runs Service in the Galtea SDK allows you to manage [runs](/concepts/product/run) for your products.
A run groups the sessions and evaluations of one launch, so a nightly job, a colleague's launch and a rerun stay separate.
This Service is exposed by the `galtea.runs` object.

<Info>
  Remember that we will be using the `galtea` object. More information [here](/sdk/api/galtea).
</Info>

## Quick Example

First, initialize the Galtea SDK:

```python theme={"system"}
galtea = Galtea(api_key="YOUR_API_KEY")
```

Open a run and put work inside it:

```python theme={"system"}
with galtea.runs.start(product_id=product_id, custom_id="ci-build-42") as run:
    print(f"Opened run #{run.ordinal} ({run.id})")

    # Every session and evaluation created inside the block joins the run.
    session = galtea.sessions.create(version_id=version_id, is_production=False)
    galtea.traces.create_batch(
        session_id=session.id,
        conversation_turns=[
            {"role": "user", "content": "What is the capital of France?"},
            {"role": "assistant", "content": "The capital of France is Paris."},
        ],
    )
    galtea.evaluations.create(session_id=session.id, metrics=[{"name": "Conversation Relevancy"}])
# Leaving the block closes the run, whether the block finished or raised.
```

Find the run again by the label you gave it:

```python theme={"system"}
run = galtea.runs.get_by_custom_id(product_id=product_id, custom_id="ci-build-42")
print(f"Run {run.id} holds {run.session_count} sessions and {run.evaluation_count} evaluations")
```

## Joining work to a run

These methods accept a `run_id`:
[`sessions.create`](/sdk/api/session/create), [`sessions.get_or_create`](/sdk/api/session/get-or-create),
[`evaluations.create`](/sdk/api/evaluation/create), [`evaluations.create_single_turn`](/sdk/api/evaluation/create-single-turn),
[`evaluations.run`](/sdk/api/evaluation/run) and [`traces.create_and_evaluate`](/sdk/api/trace/create-and-evaluate).

Inside a `with galtea.runs.start(...)` block you can leave `run_id` out: the SDK reads the open run
from the block. An explicit `run_id` always wins over the block.

<Note>
  You only close a run you opened yourself. A run that a launch opened ends on its own when its last
  launch ends, so `close()` finds it already ended and refuses it.
</Note>

## Service Methods

* [Start Run](/sdk/api/run/start)
* [List Runs](/sdk/api/run/list)
* [Get Run](/sdk/api/run/get)
* [Get Run By Custom ID](/sdk/api/run/get-by-custom-id)
* [Close Run](/sdk/api/run/close)
* [Delete Run](/sdk/api/run/delete)

## Related

<Card title="Run" icon="rocket" iconType="solid" href="/concepts/product/run">
  The record of one evaluation launch
</Card>